ELECTRONIC SEARCH BASICS  

Contents and Structure  

Electronic research resources can be structured in a number of ways—they vary from full text databases, to indexes, to something in between. Fastcase is a full text database enhanced with artificial intelligence to make searches more logical and relevant with no additional steps necessitated by the user.  Fastcase also has an important searching capability—you can retrieve documents by citation or document title (including partial title, such as party names) using the Type-Ahead function. Type-Ahead suggests the document you are looking for as you enter the portion of a document title or citation into the search bar.

You could find Marbury v. Madison, 5 U.S. 137 (1803), for example, by simply entering a portion of the party names, specifically Marb, then selecting the appropriate case from the list of documents suggested by the search logic.

Linguistic Challenges  

One of the unique challenges of electronic research, particularly with respect to full text searching, is  that as a researcher, you are generally searching for a concept rather than a particular phrase or set of  words. But full text searches are based on words, not concepts. As a result, it takes some practice—trial  and error is best—to translate concepts into effective search terms. It helps to keep two specific issues  in mind: synonymy and ambiguity.  

Synonymy means that there are many words that can express the same concept. Thus, in order to craft a comprehensive search, take a moment to think about various ways to express the concept you are researching.  

Ambiguity is the other side of the coin: any given set of words can also express more than one concept.  There are a number of ways to deal with ambiguity in the research process. For example, try adding  more search terms to eliminate false positives or irrelevant search results. There may also be other ways  to narrow down results, for example by focusing on a particular jurisdiction or time period. Fastcase  gives you powerful tools to sort and filter results in a number of ways including relevance, decision date,  case name, jurisdiction, and court hierarchy.  

Search Methods  

There are two basic approaches to searching for electronic information: searching by subject or by  keyword. One familiar example of a subject-based search is the traditional library card catalog. In the  legal research arena, West Thompson’s Key Number system is a commonly used subject-based search  tool. When it comes to searching by keyword, it is not difficult to think of an example: Google’s internet  search made keyword searching a part of everyday life.  

Subject Approach:  

Subject-based searches rely on a top-down system of categorization. Someone (or something, like a  software program) must define subject categories and decide how to organize data into those  categories. This type of system has some clear advantages and disadvantages.  

On the plus side, some important work has already been done for you. Someone has already digested  the electronic data and organized it by subject. As a result you may be able to locate a resource quickly  that would have been difficult to find any other way. For example, if you were searching a database of  book titles for a novel set in Spain, you may very well find “The Blind Man of Seville” by Robert Wilson  since the title contains a reference to the Spanish city of Seville. It is unlikely, however, that you will come across “The Shadow of the Wind” by Carlos Ruiz Zafon, set in Barcelona. In this example, a subject based index would be helpful.  

On the other hand, with subject-based searches, you are inherently relying on someone else’s system of  categorization. This introduces the possibility that there are errors in the way that the information has  been categorized or that the system of categorization is counter to the way you would organize the  same information in your own mind.  

Keyword Approach:  

Keyword searches work a bit differently. When you search by keyword, you are searching for any  instance of a search term in the electronic data. Depending on the way the electronic data is indexed,  this may mean that you can search for terms within certain fields such as title, subject, author, etc. or  you may be able to search the entire text (or full text) of the electronic data.  

Searching by keyword has the advantage of being both precise and flexible: you can search for all  electronic data containing the term(s) that you specify and there is no need to rely on a

pre-determined system of categorization. This advantage can also be seen as a disadvantage because it means you have to do a little more work. You must select keywords that accurately describe the concept or topic that  you want to research. While selecting appropriate keywords is not difficult, it can take some practice.  

Fastcase was designed using the latter-approach: keyword searching. In Fastcase, the full text of our  case law, statutes, regulations, court rules, and constitutions are all fully searchable by keyword; these  resources have not been assigned subject categories. You will find suggestions on how to construct  effective keyword searches in Chapters 2 and 4.  

 BOOLEAN SEARCHING  

By now most of us have experienced Boolean search logic in some form. Many online databases incorporate Boolean search logic, including traditional legal research services like LexisNexis and  Westlaw. Fastcase supports Boolean search logic as well. The beauty of Boolean search logic is that it  allows you to combine multiple search terms together in ways that can help you more precisely express  the concept or the topic you wish to research.  

Boolean Operators  

Boolean search logic is accomplished using a series of symbols or operators. Fastcase’s search protocol  uses the eight common Boolean operators described below:

Order of Operations  

Every search engine has its own default order of operations, the order in which it processes a query with multiple Boolean operators if you do not explicitly set the order using parentheses.

ORGANIZATION OF MATERIALS  

On Fastcase, documents are organized into categories according to the content type. There is a category for case law, one for legislative materials, etc. Currently Fastcase contains the following 5 types of primary law and secondary source documents:

Browse Libraries

Browse libraries is a function in Fastcase 7 that allows you to explore content by jurisdiction in an index style fashion.

You can access the library to browse for any jurisdiction by clicking the “Browse Libraries” button below the search bar on the main page of Fastcase 7.

The initial browse menu that will open is for your default jurisdiction. This is where either (1) you are located based on your IP address or (2) where the entity providing your Fastcase access is located. Change the browse menu jurisdiction using the filter bar in the top left of the browse menu that says “Change jurisdiction by typing.”

Content is organized by category, as discussed earlier. Every source within a content category can have as many as three icons. Hovering over each icon will tell you what it does.

Magnifying glass icon allows you to apply a source within a content category as a filter in order to run a search.

“Information Icon” looks like an “I” inside of a circle and tells you how frequently materials are being updated.

“Outline Icon” looks like three horizontal lines and allows you to browse anything with a table of contents (e.g. statutory codifications) to get more details about the items making up each organizational section.

Outline View

To explore the outline view of any eligible sources, look for the outline icon noted above to the left of the source name. Click that icon and the various versions of the source will open.

Click on the desired version and it will open the outline for that version of that source. Click each organization section to open the contents of that section. Click a volume to get the titles within, a title to get the chapters within. Rinse and repeat.

Tip: Outline View is a great way for students who are not yet comfortable searching for statutes to still be able to find a statute simply by scanning the statute sections—and section numbers—using Outline View.

Batch Downloading

In Outline View, you can click the download icon (looks like a down arrow over an em dash) to the left of any section within a source with an outline view to download the entirety of that section as a single PDF, up to 500 pages at a time.

Opening a Document Section from Outline View

As noted earlier, a student not yet comfortable searching for a legal document may first find it easier to locate by browsing. To open a document from Outline View, simply open up each section within the Outline of the chosen source until you arrive at the blue hyperlinked documents.

Clicking the blue hyperlinked document name from the Outline view of the chosen source will open the text of that document. The text of the document opens on the right while the outline view remains open on the left.

You can then click over to other documents within the Outline View by either clicking the blue hyperlinked document title(s) on the left or using the doc arrows above the text of the presently open document section on the right.

Tip: Once a student is able to locate and open a document via Outline View, the title of that document—particularly with statute sections—is a good indicator of how to search for it within Fastcase. For example, in the above image Section 106.020 of the Oregon Revised Statutes was opened. The title of that statute section is ORS 106.020. Running a search for ORS 106.020 gets that statute section at the top of your search results.

History

Fastcase stores a history of (1) documents opened and (2) searches conducted, with no expiration date. To access the history, click the icon shaped like a clock in the top right of any page on Fastcase 7.

Search history is the tab on the far left. Search history is organized in reverse chronological order. You can click the star to the right of any search in your search history to save it. Once a search is saved, it moves from the far left tab under the clock icon to the tab that is second from the right, appropriately called “saved searches.”

Document history is the second tab from the left under the clock icon and, like search history, is organized in reverse chronological order. To the left of each document in the document history, you will find two icons. The icon on the far left is the bookmark icon. You bookmark a document to save it. Once a document is bookmarked, it moves from the second tab from the left to the center tab called “bookmarks.”

You can also build an export queue—download many documents as a single document or zip file—from the document history using the icon to the right of the bookmark.

Tip: Icons in Fastcase 7 fill in solid with color when their function is applied.

SEARCHING IN FASTCASE 7

Fastcase 7 has a single search bar that can run searches for keywords, natural language phrases, or citations without a need to specify the types of searches being conducted.

Locating a Document Using a Citation

The easiest way to locate a document when you know the complete citation of the document is to use Type Ahead. Plessy v. Ferguson is reported at 163 U.S. 537, so I enter 163 U.S. 537 into the search bar and the case I seek populates in the suggested documents menu below said search bar. Click the case name to open it.

You can also run the search for the citation, in the event you are unable to use Type-Ahead. The search logic will recognize the citation as a citation and place it in quotes. Your desired case then appears at the top of the search results.

Note: When you are searching for a document via a citation sans Type-Ahead, the sought after document may not always be number 1 on the list, but will usually be in the top 5 of your search results.

Natural Language Searching

 Natural language searching is best thought of as being searching posed as if one is asking a question. If you were a law clerk for a certain cartoon judge, a natural language search might be Who Framed Roger Rabbit? If you are hungry, but strapped for cash, you might search for What is the Cost of a Free Lunch?

The problem with natural language searching is that most legal research platforms—including Fastcase 7— use an implied AND. An implied AND simply means that when you are running a natural language search, the most relevant results—as shown above—will be those results with your desired natural language phrase, but more irrelevant results will simply be in your search results because the implied AND search logic mandates that some documents make it into the search results simply due to containing all the words making up your natural language search phrase.

The Roger Rabbit example shows how natural language searches can go from easy and relevant to irrelevant due to the implied AND.

Result #3 contains the exact phrase from the natural language search, while result #6 made the list for containing who AND rabbit AND framed AND roger.

Keyword Searching

Keyword searching is the most accurate way of searching within full-text databases like Fastcase 7. Basic keyword searching involves entering various words or phrases you seek to research into the search bar and running your search. If I was looking for a case or cases about cows with a “vicious propensity” to attack others, I would simply enter something like: cow attacked woman injured “vicious propensity.”

Tip: Phrases can be searched within quotations in order to ensure that no words or characters are allowed between the words within said phrase. Quotes also ensure that words in a phrase appear in the desired order within search results. Summary judgment has a much different meaning than judgment summary.

A keyword search—like a natural language search—will only populate search results containing all the keywords in a query. Use OR between keywords or phrases in order to specify any number of keywords are optional. We discuss this in more detail below.

Searching with Keywords Involving Boolean Operators

Many people confuse keyword searches with boolean operator searches, so it is important to know the difference. First, encourage yourself and your students to remove the phrase “boolean operator search” or “boolean search” from one’s vocabulary.

Why? Boolean operators are not their own kind of search or even searchable on their own. Boolean operators are simply a limited universe of connecting terms and functional characters that establish relationships between keywords in order to make a search more tailored and thus search results more relevant.

Thus the better phrase to remember is keyword searching with boolean operators. Now many students will hear the word boolean and initially—as I did—feel intimidated. However, boolean is not difficult. To prove it, I will lay out how to master it in five easy steps.

Step One: Only three terms matter in the universe of boolean operators: AND, OR, NOT. As such, technically only two terms matter, as the absence of boolean operators in a keyword search means there is the implied AND between keywords.

Use NOT to exclude whatever follows the word NOT. If you want to know about the murder she did not write, your search could be murder not wrote. This find documents  containing the word murder but not containing the word wrote.

Tip: Use one NOT per excluded word. The Fastcase 7 search logic handles the rest. Example might be: murder not wrote not published which becomes: murder not (wrote or published). Students can learn the elaborate “polished” version of the query, or just remember one NOT per excluded word and let the AI handle the rest.

Use OR in search logic as you use OR in normal english. It is used to specify satisfaction with a document in your search results containing but one of two or more keywords. OR can also be used to account for variations of terminology, phrasing, or even well known spelling errors by judges.

If you are looking up old lawsuits involving Katy Perry, you might not be finding your sought out case(s), only to discover Katy Perry used to be Katy Hudson. As such, your search might now be: “katy hudson” or “katy perry.”

If you remember when Kesha spelled her name with a dollar sign as Ke$ha, you may also remember that a few judges referred to her exclusively with the dollar sign spelling—throughout the entire opinion. As such, you might look up cases involving Kesha—now dollar sign free—as: kesha or ke$ha.

Step Two: Parentheses are the Marie Kondo of Boolean Operators: Here to bring you joy and sort similar things in similar locations.

Put similar things in the same parentheses in order to do more things with your search at one time. Do one at a time. No multitasking with parentheses. Open one. Close it. Open another. Close it.

If we return to the earlier example about the angry cow, let us do a modified hypothetical.

Your client has been attacked by angry livestock after she prodded the neighbor’s cow with a blunt broom to get it off their porch and you intend to get them the compensatory damages they deserve. But you first need the perfect case! Parentheses to the rescue.

• You know that the judge may refer to cows as also cattle or livestock. Because Marie Kondo is designing our search query, similar things go in similar locations e.g. the same set of parentheses. Open the parentheses. Add cattle or cow or livestock. Close parentheses
  • Current draft search: (cattle or cow or livestock)

• You know that the neighbor’s cow attacked your client. You brainstorm—maybe even grab a dictionary—to see all the variations of attack that a cow could do. Open a new parentheses. Add attacked or assaulted or trampled. Close parentheses.
  • Current draft search: (cattle or cow or livestock) (attacked or assaulted or trampled)

• Finally, you have to prove the cow was angry, as no Judge would believe a cute calm cow would stomp on your client. So you need the case to also show that even cute cows become bitter bulls when prodded with household brooms. You open parentheses. Enter angered or enraged or angry or mad. Close parentheses.
  • Current draft search: (cattle or cow or livestock) (attacked or assaulted or trampled) (angered or enraged or angry or mad)

Tip: Students should feel no shame in resorting to larger search engines like Google or Merriam Webster’s Dictionary to locate appropriate synonyms for common words.

Step Three: 10s. 10s. 10s Across the Board. Are you in the proximity?

A proximity connector finds relationships between keywords. If you search summary judgment without a proximity connector, the word summary and judgment will both appear in all documents in the search results, but may be paragraphs apart. A proximity connector—like a good hug—keeps things together by limiting distance between two things that desire to be close. Proximity connectors, as discussed earlier, use w/n between two keywords, phrases or parentheses with n being replaced by a number that represents the maximum number of words allowed to separate the parts of the search query joined by the proximity connector.

As a best practice, proximity connectors should never be less than 10, UNLESS you are searching for a document title, a name, or variations of a phrase with a definable length. This is because proximity connectors must account for “noise words”—words that are necessary to complete a sentence, but that provide no substantive meaning—which there are seven of note: A, An, The, For, But, Nor, So. 

Step Four: Extend at the end, maybe even just to amend.

The boolean operator known as the “wildcard” is the most frequently—and most incorrectly—used of the operator universe. A wildcard is easy to use. You place a * or ! at the end of a word, and the system includes every variation of the word available; using up to 128 independent endings attached to the root of the wildcard word preceding the wildcard operator.

So, to put it simply, corp* would look for corporate, corporation, and even corporal. This can seem useful, but can also quickly balloon your results.

As such, best practice is to design the rest of your search—perhaps even running a search first for relevancy of results—before adding the ever-tempting wild card.

Returning to our working example of (cattle or cow or livestock) (attacked or assaulted or trampled) (angered or enraged or angry or mad):

• If instead of the above example, I switched attacking to attack*, I go from 385 results to 715 results.

Step 5: Have a strategy or copy mine.

I make it a practice to work every search in the following way.

1. I pose my legal research query into a single sentence. “Find a case where a cow attacked a neighbor after being enraged in response to being prodded with an object while lounging on the porch of the non-owner neighbor.”
2. I break that out into no more than three pods by omitting superfluous words. “Mad Cow attacked person after being struck by object.” I place entries for pods—even if one word— in parentheses. (MAD) (COW) (ATTACKED).
3. I look up and add synonyms to each pod. (cattle or cow or livestock) (attacked or assaulted or trampled) (angered or enraged or angry or mad)
4. I add an initial proximity connector of w/10 between the first two pods. (cattle or cow or livestock) w/10 (attacked or assaulted or trampled) (angered or enraged or angry or mad).
5. I examine results. I add, if needed, w/10 between the second and third pods. (cattle or cow or livestock) w/10 (attacked or assaulted or trampled) w/10 (angered or enraged or angry or mad).
6. If needed, I add 5 to each proximity connector and wildcard operators as necessary.
7. I enjoy the fruits of my labor with the intended case now in my grasp.

Note: Poor Susan should not prod any more cows.

Narrowing a Search

As explained above, a good search helps narrow your search results generally, however, it is important to make sure you are getting your legal documents from the correct jurisdiction as law is only precedential if it is primary mandatory authority.

The American legal system consists of law from several levels of government and various types of authority. Legal authority is divided into two major categories: they are primary and secondary authority. These types of authority affect the research process because of their relative importance.

Primary authority is the law itself, and therefore, the most desirable source of legal research. Primary authority consists of written constitutions, statutes, and court decisions. These authorities are in turn designated as either mandatory or persuasive.

Primary mandatory authority consists of the constitution, statutes, and decisions of the highest level of courts from a jurisdiction. For example, all trial and intermediate appellate courts in California must follow the California constitution, statutes, and Supreme Court of California decisions.

Primary persuasive authority consists of appellate court decisions from other jurisdictions. However, the constitutions or statutes from other jurisdictions are neither mandatory nor persuasive. Because primary authority is the most important source, the researcher should always try to locate primary authority to support the client's position.

Secondary authority is all other written expressions of the law. As a rule, secondary authority explains or expounds upon primary authority. It is generally used if there is no primary authority to support or explain the legal issue. Examples of secondary authority include treatises, law review or bar journal articles, and legal encyclopedias.

Applying a Jurisdictional Filter

In Fastcase 7, you generally apply a filter—after clicking Jurisdictions & Sources to the right of the search bar to opens the filters menu–by clicking the state or federal circuit on the left THEN the content type on the right.

Each content category on the right has a small arrow to the left of the name of the content category to open up all the individuals sources or sub-jurisdictions (e.g. individual federal district courts) within each content category.

SAMPLE RESEARCH EXERCISES 

Below you will find sample research exercises that you can use to help your students familiarize themselves with Fastcase. We have included suggested research paths for each exercise, but these are only suggestions. There are many additional research paths that may yield the same or similar results.  

BASIC  

1. Find this case using only the party names: Gideon v. Wainwright  

1. 1. Answer: Enter Gideon w/5 Wainwright into the search bar.
      1. Case is the #2 search result.

Tip: This is a perfect scenario where a researcher could skip the boolean operator and use Type Ahead. Entering Gideon into the search bar makes the Type-Ahead function populate the case you are looking for as a suggested document below the search bar. Click the case name, and it will open.  

2. How many votes are required to amend the Oregon Constitution?  

Answer:

• Search amend* w/10 constitution in the Oregon Constitution.
  • OR. Const. Art XVII Sec. 1 Method of Amending Constitution requires a majority of the legislature to approve an amendment.

3. What is the maximum size of a Grand Jury according to the Federal Rules of Criminal Procedure?  

Answer:

• From the main page of Fastcase 7, click “Browse Libraries” below the search bar.
• With the browse menu open, select Federal Rules of Criminal Procedure under the column of Cases and Court Materials.
• Select the 2017 edition of the Federal Rules of Criminal Procedure to open the outline view.
• Once the outline view opens, select Title III: The Grand Jury, the Indictment, and the Information.
• Click the Title opens the rules in that title. Select Rule 6: The Grand Jury
• Rule 6(a)(1) says that a jury can be no less than 16 and no more than 23 jurors.

 4. How would you search for decisions (any jurisdiction) citing 42 U.S.C. § 1983? 

Answer:

• From the main page of Fastcase 7, select Jurisdictions and Sources to the right of the search bar.
• Once the pop-out opens, check the boxes for cases on the right side.
• Click Save and Continue in the top right to close the pop-out and apply a filter for all state and federal cases.
• Enter the search query of "42 u.s.c. sec. 1983" or "42 u.s.c. § 1983"
• Using both Sec. and § ensures you catch all variations of the statute citation, since judges are     never uniform on this front.
• Ensure you have 318,000 results or more.